Alexander Matthew Payne

It's a bit maddening that they didn't publish equations in this paper. I *think* for a two model comparison of reference A to model B: 1. For models A and...

Are you connected with them at all? I am not. I think I'd be happy to take a stab at implementing but have no confidence in doing it the fastest...

Looks like there might be an implementation from the Baker Lab: https://github.com/hiranumn/DeepAccNet/blob/21c1775f841bea685b17c688a64da733e5d55cf7/deepAccNet_noPyRosetta/model.py https://github.com/hiranumn/DeepAccNet/blob/21c1775f841bea685b17c688a64da733e5d55cf7/deepAccNet/model.py

> Sorry here by "interface" I meant that one could feasibly just make a "converter" that takes an MDAnalysis AtomGroup and uses their Python API to run the actual calculation....

One thing I'm not sure how to standardize (and would like to dig into the other implementations) is the alignment for the two proteins. I'm thinking we'd want users to...

> From reading it I have come to the same interpretation as you though. Best, Dirk Thanks for double-checking @dirkwalther ! > Alphafold source code contains the implementation for the...

@Ieremie this is fantastic, thank you!

@Aryaman-Chaudhri thanks for linking this! I haven't had a chance to check this out yet and haven't thought about this in a while but this should be nice to have....

alchemy CI is hitting errors because of this issue, should be resolved soon: https://github.com/conda-forge/openmmforcefields-feedstock/issues/25